Download Lithosphere structure underneath the Tibetan Plateau inferred from

Available online at www.sciencedirect.com
Earth and Planetary Science Letters 267 (2008) 276 – 289
www.elsevier.com/locate/epsl
Lithosphere structure underneath the Tibetan Plateau inferred
from elevation, gravity and geoid anomalies
Ivone Jiménez-Munt a,⁎, Manel Fernàndez a , Jaume Vergés a , John P. Platt b
a
Group of Dynamics of the Lithosphere, Institute of Earth Sciences ‘J. Almera’- CSIC, Sole Sabaris s/n, 08028 Barcelona, Spain
b
Department of Earth Sciences, University of Southern California, Los Angeles, CA 90089-0740, USA
Received 30 January 2007; received in revised form 22 November 2007; accepted 23 November 2007
Available online 8 December 2007
Editor: C.P. Jaupart
Abstract
The Tibetan Plateau is the product of crustal thickening caused by collision between India and Asia. Plate tectonic reconstructions suggest
continuous northward movement of the Indian plate relative to stable Eurasia at nearly 50 mm/yr for the last 50 My. The plateau is now at ~ 5 km
elevation with steep topographic gradients across the southern and northern margins. These steep topographic gradients are also related to large
lateral variations in the geoid and gravity anomalies. In a SSW to NNE cross section, the Bouguer gravity anomaly decreases over a distance of
500 km from about 0 mGal in the India plate to ~ −500 mGal on the Tibetan Plateau. The geoid anomaly also presents steep gradients on both the
Himalayan front and the northern margin, reaching values between 20–30 m on the plateau, suggesting a pronounced thinning of the lithospheric
mantle. Uplift late in the tectonic evolution of the plateau, the widespread extension, and the associated magmatism have been attributed to
convective removal of the lower part of lithospheric mantle and its replacement by hotter and lighter asthenosphere. Here we present a twodimensional lithospheric thermal and density model along a transect from the Indian plate to Asia, crossing the Himalaya front and the Tibetan
Plateau. The model is based on the assumption of local isostatic equilibrium, and is constrained by the topography, gravity and geoid anomalies
and by thermal data within the crust. Our results suggest that the height of the Tibetan Plateau is compensated by thick crust in the south and by
hot upper mantle to the north. The Tibetan Plateau as a whole cannot be supported isostatically only by thickened crust; a thin and hot lithosphere
beneath the northern plateau is required to explain the high topography, gravity, geoid and crustal temperatures. The lithosphere reaches a
maximum depth of ~260 km beneath the southern Plateau, and thins abruptly northward to ~ 100 km under the central and northern Plateau. The
lithosphere depth increases again beneath the Qaidam basin and the Qilian Shan to ~ 160 km.
© 2007 Elsevier B.V. All rights reserved.
Keywords: modelling; asthenosphere; lithosphere removal; uplift; potential fields
1. Introduction
The deep structure of the India–Asia collision and the
Tibetan Plateau has been a subject of debate for a long time. The
Cenozoic collision between the Indian and Asian continents
formed the Tibetan Plateau, beginning at about 55 Ma with an
average 50 km/My N–S convergence. Since this time, at least
⁎ Corresponding author.
E-mail addresses: [email protected] (I. Jiménez-Munt),
[email protected] (M. Fernàndez), [email protected] (J. Vergés),
[email protected] (J.P. Platt).
0012-821X/$ - see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.epsl.2007.11.045
1400 km of convergence has been absorbed within the
Himalayan–Tibetan orogen (Yin and Harrison, 2000) by a
combination of several mechanisms. A wide variety of
hypotheses have been proposed for Tibetan tectonic evolution,
such as crustal shortening and thickening by pure shear during
the Mesozoic and Cenozoic (Murphy et al., 1997) or entirely
during the Cenozoic (Houseman and England, 1993), delamination or convective removal of tectonically thickened lithospheric mantle (England and Houseman, 1989; Molnar et al.,
1993; Platt and England, 1994; Jiménez-Munt and Platt, 2006),
block extrusion along major strike-slip faults (Tapponnier et al.,
2001), and thrusting of the Indian mantle lithosphere under Asia
I. Jiménez-Munt et al. / Earth and Planetary Science Letters 267 (2008) 276–289
(Owens and Zandt, 1997). Some studies have proposed
combinations of more than one of these mechanisms for thickening and uplift of the plateau. The seismicity indicates that
normal and strike-slip faulting occur at elevations above 4 km,
with consistent east-west orientation of crustal extension
(Molnar and Lyon-Caen, 1989). Thrust faulting in regions
with lower elevations continues to build a thicker crust and to
elevate the margins of the plateau.
The 2-D lithospheric model presented in this study traverses
2300 km along a SSW–NNE trending transect crossing, from
India to Asia, the eastern region of the Himalaya, Tibetan
Plateau, Qaidam Basin, Qilian Mountains and Beishan. This
region has experienced a long history of accretion, arc and
microcontinental collisions, and associated tectonism during the
Phanerozoic, resulting in a wide variety of geological composition and structure. In the following paragraphs we review
the likely crustal structure and tectonic evolution of these
terranes in turn.
Peninsula India south of the Himalaya is largely composed of
the Precambrian Indian shield, which is itself a collage of
cratons and intervening mobile belts assembled between midArchean and neo-Proterozoic time (Naqvi and Rogers, 1987),
and carved into a peninsula by south-tapering passive
continental margins that are covered by a thin veneer of
Phanerozoic sediments. The Himalayan fold and thrust belt
marks the northern collisional boundary of India with Tibet, and
lies between the Indian shield and the Indus–Tsangpo Suture
(Fig. 1). It consists of three major tectonic slices, each of which
shows extensive internal imbrication and thrusting: the Lesser
Himalaya, Greater Himalaya and Northern or Tethyan Himalaya. They are all considered to have been formed from the
north-facing Himalayan passive continental margin, which
developed from middle Proterozoic to Cretaceous times (Yin
and Harrison, 2000). The Greater Himalaya in particular shows
extensive high-grade metamorphism, partial melting, and
intrusion by anatectic granites. It was exhumed by normal-
277
sense motion along the South Tibetan detachment, which
separates it from the Tethyan Himalaya. The Himalaya as a
whole is separated from the Tibetan Plateau proper by the
Indus–Tsangpo Suture, which marks the original boundary
between the Indian and Asia continental masses.
In the eastern part of the Tibetan Plateau, which is the widest,
five major crustal blocks can be distinguished: the Lhasa,
Qiangtang, Songpan–Ganzi, Kunlun–Qaidam and Qilian Shan
terranes (Fig. 1). Each of these represents a separate arc,
accretionary complex, or microcontinental fragment that was
progressively accreted to the southern Asian margin during the
Phanerozoic.
The Lhasa terrane is bounded by the Indus–Tsangpo Suture
and the Bangong–Nujiang Suture, and it collided with
Qiangtang to the north in the late Jurassic (Dewey et al.,
1988). The Lhasa terrane is about 300 km wide at longitude
91°E and narrows westwards. Sedimentary strata consist of a
sequence of Ordovician and Carboniferous to Triassic shallow
marine clastic sediments, and the basement is mid-Proterozoic
to early Cambrian in age (Yin and Harrison, 2000). The Lhasa
terrane probably rifted away from northern Gondwana during
Triassic time and collided with the southern margin of the
Qiangtang terrane in Late Jurassic–middle Cretaceous time
(Dewey et al., 1988). Along its southern margin, the Paleozoic
and Mesozoic sequences are intruded by the dominantly
Cretaceous to Tertiary Gangdese batholith belt, which formed
an Andean-style continental margin arc above the subducting
Neotethyan oceanic slab.
The Qiangtang terrane lies between the Bangong–Nujiang
and the Jinsha Sutures, and is about 500 km wide in central
Tibet. It rifted off of northern Gondwana during late Paleozoic
time and collided with Eurasia in late Triassic–early Jurassic
time. It comprises a high-pressure/low temperature metamorphic complex of Early Triassic age lying structurally
beneath Late Paleozoic to Jurassic shallow marine carbonate
Fig. 1. Tectonic map of the Tibetan plateau and surrounding regions. Colour shades show elevation. Red line shows the position of the modelled profile. ATF: Altyn
Tagh Fault, HFF: Himalayan Frontal Fault, MBT: Main Boundary Thrust of Himalaya, MCT: Main Central Thrust, ITS: Indus–Tsangpo Suture, BNS: Bangong–
Nujiang Suture, JS: Jinsha Suture, KF: Kunlun Fault, NBT: North Border Thrust.
278
I. Jiménez-Munt et al. / Earth and Planetary Science Letters 267 (2008) 276–289
rocks interbedded with terrestrial clastic and volcaniclastic
strata (Yin and Harrison, 2000).
The Songpan–Ganzi terrane is a triangular tectonic element
and in the central Tibet it extends in a narrow belt about 100 km
wide, between the Qiangtang terrane and the Kunlun–Qaidam
terrane. It is characterized by a thick sequence of deep marine
Triassic strata, commonly referred as the Songpan–Ganzi flysch
complex.
Northward there is the eastern Kunlun–Qaidam terrane
bounded to the north by the southern Qilian Suture. In the south,
it is dominated by a broad Early Paleozoic arc, on which a
younger and narrower Late Permian to Triassic arc was
superposed (Yin and Harrison, 2000). The northern part is
mostly occupied by the Qaidam basim.
The Qilian Shan is the northern terrane of the Tibetan
Plateau, and it consists of complexly deformed early Paleozoic
arcs, which were developed at the southern margin of the North
China craton before it was offset by the Altyn Tagh fault in the
Cenozoic (Yin and Harrison, 2000).
Recent geophysical and geological studies indicate that the
Tibetan Plateau has a heterogeneous structure in the crust and
lithospheric mantle (Nelson et al., 1996; Owens and Zandt,
1997; Hacker et al., 2000; Wei et al., 2001) and widespread
post-collisional magmatism (Turner et al., 1993, 1996; Chung
et al., 2005, and references therein). High-potassium magmas
mixed with a mantle component have erupted on the Tibetan
Plateau since ~ 45 Ma, with major pulses of activity during late
Eocene–Oligocene time in the Qiangtang terrane and during the
Miocene–Quaternary in the Songpan–Ganzi and Lhasa terranes
(Turner et al., 1993). Widespread geothermal activity and
several independent geophysical and petrological data sets
indicate that the crust of the Tibetan Plateau is abnormally hot,
although the debate persists whether partial melt is pervasive
throughout the plateau or restricted to the extensional grabens
(Nelson et al., 1996; Owens and Zandt, 1997; Alsdorf and
Nelson, 1999; Hacker et al., 2000; Wei et al., 2001; Kind et al.,
2002). Seismic reflection and magnetotelluric profiling in
southern Tibet have revealed the presence of fluid concentration
at 15 km depth (Brown et al., 1996). The shallow Curie
isotherm suggested by the pronounced low satellite magnetic
anomaly associated with the Tibetan Plateau (Alsdorf and
Nelson, 1999), active volcanism widespread across the
Qiangtang block, and magnetotelluric results (Wei et al.,
2001), all support the hypothesis that melt exists within the
crust of central and northern Tibet.
The structure beneath the Tibetan Plateau started to be
delineated with the results of the Sino–American PASSCAL
broadband experiment (Owens and Zandt, 1997; McNamara
et al., 1997), the international and multidisciplinary INDEPTH
experiments (Nelson et al., 1996; Zhao et al., 2001), and the
series of Sino–French seismic studies (Hirn et al., 1995; Galvé
et al., 2002; Jiang et al., 2006). A large diversity of geophysical
methods has been used to study the lithosphere of the Himalaya
and Tibet regions, such as deep seismic (Hauck et al., 1998;
Galvé et al., 2002; Haines et al., 2003; Meissner et al., 2004;
Jiang et al., 2006), seismic tomography (McNamara et al., 1997;
Villaseñor et al., 2001), magnetotellurics (Wei et al., 2001;
Unsworth et al., 2004), gravimetry (Jin et al., 1996; Braitenberg
et al, 2000), seismology (Mitra et al., 2005) and geothermics
(Chung et al., 2005). These geophysical studies mostly defined
the crustal structure whereas the mantle lithosphere remained
little constrained. However, most of those data have been
interpreted separately and the structure and geodynamics of the
mantle beneath the plateau have been subject to different
interpretations. This led us to the application of an integrated
lithospheric modelling, based on the combined interpretation of
geophysical (topography, gravity, geoid and thermal) and
geological data.
In this paper we present a 2-D lithospheric thermal and
density model along a SSW–NNE trending transect crossing the
eastern region of the Himalaya and Tibetan Plateau from India
to Asia. The model is based on the assumption of local isostatic
equilibrium and constrained by the topography and by gravity
and geoid anomalies. The main objective of this study is to
model the lithosphere structure beneath the Tibetan Plateau and
its margins, and its consequences in terms of the geodynamic
evolution.
2. Method and data analysis
The numerical model used is based on the integration of
regional elevation, gravity, geoid and thermal data together with
available seismic data. The combined interpretation of these
data provides information on the density and temperature
distribution at different depth ranges. Whereas short-wavelength gravity anomalies and surface heat flow variations are
sensitive to shallow structures, elevation, geoid height and longwavelength heat flow variations are more sensitive to deep
structures. Seismic data provide valuable information on both
the geometry of crustal layers and the depth distribution of
P-wave velocities.
Geopotential, lithostatic and heat transport equations were
solved simultaneously using a finite element code based on
Zeyen and Fernàndez (1994). The lithospheric section is
divided into a number of bodies with different material
properties: thermal conductivity, radiogenic heat production
(which can vary with depth), and density (which can vary with
pressure and temperature). The top and bottom of the model
correspond to the Earth's surface and the lithosphere-asthenosphere boundary, respectively. The density of the lithospheric
mantle is assumed to be temperature dependent according to
ρm = ρa(1 + α·(Ta − T(z)), where ρa is the density of the asthenosphere, α is the thermal expansion coefficient, and Ta is the
temperature of the asthenosphere. Temperature distribution and
surface heat flow are calculated by solving the steady-state heat
transport equation assuming no heat flow across the lateral
boundaries and fixed temperatures at the top and bottom of the
model of 5 °C and 1350 °C, respectively. The Bouguer gravity
anomaly is calculated using Talwani's 2D algorithm (Talwani
et al., 1959) for each triangular element of the finite element
mesh allowing for density variations depending on pressure,
temperature, and lithology. The geoid anomaly is calculated
using an algorithm based on the resolution of the gravity
potential for a rectangular prism, which is formed by adjacent
I. Jiménez-Munt et al. / Earth and Planetary Science Letters 267 (2008) 276–289
triangular elements and extends to infinity in a direction
perpendicular to the strike of the lithospheric section (Zeyen
et al., 2005). Elevation is calculated for every column of the
mesh by comparing its buoyancy with that corresponding to the
average mid-oceanic ridge column under the assumption of
local isostasy and following Lachenbruch and Morgan (1990).
The depth of isostatic compensation to calculate elevation, and
gravity and geoid anomalies is taken at the maximum depth
reached by the lithospheric mantle along the transect. The space
between this depth and the base of the model is filled with
asthenospheric material with a constant density. The resulting
elevation, Bouguer anomaly, geoid height variation, and surface
heat flow or crustal temperatures are compared with the
measured values. An initial crustal model is constructed by
compiling available seismic data and other geophysical studies,
and then the geometries of the lithosphere–asthenosphere
boundary and crust are modified until the best fit is obtained.
Clearly, the degree of freedom in modifying the geometry of the
crustal layers depends on the availability and quality of seismic
data. This modelling technique has been applied in different
geodynamic settings to determine the crustal and lithospheric
structure, including the Pyrenees (Zeyen and Fernàndez, 1994),
279
the High-Middle Atlas (Zeyen et al., 2005), and the Vøring and
the SW Iberian margins (Fernàndez et al., 2004a,b).
The selected 2D lithosphere profile crosses the easternmost
part of the Tibetan Plateau (line from Figs. 1 and 2), from the
Indian shield (87°E, 23.46°N) to the southern Qaidam Basin
(94°E, 36°N) and northeastward to the Beishan Basin (100°E,
41.5°N). It follows the major geophysical studies done in the
plateau: INDEPTH I and II (Hauck et al., 1998), PASSCAL
(Unsworth et al., 2004), and the Golmud–Ejn transect (Rui
et al., 1999). Our main interest was the Tibetan Plateau and its
margins, although we extended the profiles ~ 400 km south into
the Indian plate and north into the Beishan Basin to avoid
border effects and to model lithosphere mantle depths. Therefore, the modelling approach searches a good fit to gravity,
geoid, topography and thermal data along the 2300 km long
profile.
The topographic data we used is taken from GINA Global
Topo Data (http://www.gina.alaska.edu), with values in a 30 s grid
spacing (Fig. 2a). The plateau is now at an average elevation of
5 km with steep topographic gradients across the southern and
northern margins. At the Himalayan front, the elevation increases
from 100–300 m in the Indian plate to ~5000 m in the plateau in
Fig. 2. Data: a) topography; b) geoid (Lemoine et al., 1998); c) Bouguer gravity calculated from global free air (Sandwell and Smith, 1997) with the 3D topographic
correction (Fullea et al., submitted for publication); d) surface heat flow measurements from the global data set (Pollack et al., 1993) and from the Golmud–Ejn transect
between Qaidam and Beishan basin (Rui et al., 1999). Grey line shows the position of the modelled profile. Note that d) is at a larger scale than the other figures.
280
I. Jiménez-Munt et al. / Earth and Planetary Science Letters 267 (2008) 276–289
less than 200 km horizontally. Also, on the northern margin, north
of the Qaidam Basin (with a relative topographic minimum,
2700 m) and the Qilian Shan (~4500 m), the elevation presents a
steep gradient decreasing below 1000 m.
Geoid height (Fig. 2b) is taken from the EGM96 global
model (Lemoine et al., 1998). In order to avoid effects of
sublithospheric density variations on the geoid, we have
removed the geoid signature corresponding to the spherical
harmonics developed until degree and order 8. The geoid
anomaly also presents a steep gradient at the Himalayan front
and the northern margin. The values over the Himalaya are
between 30 and 35 m, and they smoothly decrease to 10 m
towards the NE on the plateau. Farther to the NE, negative geoid
anomaly values are recorded in the Qaidam Basin (− 7 m) and in
the Beishan region (− 12 m). Negative values are also observed
west to the profile coinciding with the Tarim Basin and the
Himalayan front.
The Bouguer gravity anomaly (Fig. 2c) is calculated from the
global free air anomaly database (Sandwell and Smith, 1997) to
which a full 3D topographic correction following the methodology described by Fullea et al. (submitted for publication) has
been applied using a density reduction of 2670 kg m− 3. The
Bouguer anomaly on the Indian plate is ~ 0 mGal and it
decreases to the north down to ~ − 500 mGal on the Tibetan
Plateau, being the lowest Bouguer anomaly values on Earth.
North of the Qaidam Basin the Bouguer anomaly starts to
increase reaching − 150 mGal in China. Seismic studies
crossing the plateau found a thick crustal root underneath the
plateau, which would be in agreement with this high elevation
and low values of the Bouguer gravity anomaly.
Thermal data come from several sources. The heat flow data
(Fig. 2d) is taken from that compiled on the global data set
(Pollack et al., 1993) and those measurements done on the
Golmud–Ejn transect between Qaidam and Beishan basin (Rui
et al., 1999). Heat flow measurements in the Tibetan Plateau are
scarce and they include measurements on lake floors, water
exploration wells and oil wells. The reported values show a high
scatter probably related to active groundwater flow and
magmatism as evidenced by the presence of numerous geysers,
hot springs and volcanic activity. In spite of that, heat flow data
suggest a decreasing trend from south to north, with values of
~ 100 mW/m2 and higher in the south and around 60 mW/m2 in
the north. In addition to heat flow measurements, we also use
indirect thermal data to constrain the temperature at different
crustal and lithospheric levels. Seismic detection of the α-β
quartz transition suggests a temperature of 700 °C between
the upper and the middle (18 ± 2 km) crust under the southern
Qiangtang block and 800 °C at 32 ± 3 km depth in the northern Lhasa block (Mechie et al., 2004). Hacker et al. (2000)
also found temperatures between 800–900 °C at 0.8–1.3 GPa
(30–50 km) in the central Qiangtang from analyses on xenolith
samples. However, Galve et al. (2006) propose lower temperatures from P-wave velocity data, and depending on
composition they suggest values of ~ 500 °C and 850 °C
between 1 and 2 GPa (37 and 70 km) in the lower crust under
the Tethyan Himalayan.
3. Crustal structure
An initial crustal model based mainly on available seismic
data was constructed to start the lithosphere modelling. The
geometry of the different crustal layers was inferred from
seismic profiles where available (Hauck et al., 1998; Haines
et al., 2003; Mitra et al., 2005; Zhang and Klemperer, 2005;
Jiang et al., 2006). Crustal and upper mantle densities have been
calculated at some sites along the transect from the proposed
seismic velocities (Vp and Vs) and using the velocity–density
relation of Christensen and Mooney (1995).
In the Indian plate and the southern Tibetan Plateau we used
the results from the deep seismic reflection profiles of the
INDEPTH I and II projects (Hauck et al., 1998), and the results
by Mitra et al. (2005) from teleseismic receiver function
analysis. These works helped us to fix the thicknesses of the
Indian foreland basin, upper and lower crust, and the geometry
of the main Himalayan thrust, Lesser and Greater Himalayan
units, the Tethyan Himalaya, and the Gangdese batholith
(Fig. 3). From these studies we deduced a crustal thickness on
the Indian plate of 40 km, which increases to 75–80 km beneath
the northern Tethyan Himalaya and the southern Lhasa terrane.
The INDEPTH III project provides valuable information
about the crustal thickness and density/depth variations in the
northern Lhasa terrane from a W–E wide angle seismic profile
(Zhang and Klemperer, 2005); and in the southern Qiangtang
terrane from reflection data (Haines et al., 2003; Zhao et al.,
Fig. 3. Crustal structure along the profile inferred from geological (Yin and Harrison, 2000) and geophysical data. Bars show some moho depth constraints from several
geophysical experiments: (1) black marks (Mitra et al., 2005), (2) (Hauck et al., 1998), (3) (Zhao et al., 2001), (4) (Owens and Zandt, 1997; Meissner et al., 2004),
(5) (Haines et al., 2003), (6) (Zhang and Klemperer, 2005), (7) (Owens and Zandt, 1997; Zhao et al., 2001; Haines et al., 2003), (8) (Owens and Zandt, 1997; Zhao
et al., 2001; Kind et al., 2002), (9) (Jiang et al., 2006; Zhao et al., 2006), (10) (Rui et al., 1999), (11) (Vergne et al., 2002).
I. Jiménez-Munt et al. / Earth and Planetary Science Letters 267 (2008) 276–289
281
Table 1
Physical parameters of the different bodies used in the modelling
H
K
ρ
V and [Ref.]
1.5 exp (−z / 15)
2.2 exp (−z / 15)
2.3
2.5
Vs = 2–2.75 (Mitra et al., 2005)
Vs = 3.5–3.8 (Mitra et al., 2005)
2.5 exp (−z / 15)
2.5
2450
2700
2890
2750
2910
1.2
2.3
Vs = 2.5–3.2 (Mitra et al., 2005)
Vp = 5.2 (Hauck et al., 1998)
2.0
3.0
2650
2700
2750
2.5 exp (−z / 30)
2.2
2700
2930
Triassic-Jurassic shallow marine carbonates
interbedded with terrestrial clastic and
volcaniclastic strata
Low to high grade metamorphosed
melange complexes
1.2
2.3
2500
2600
density (Haines et al., 2003)
Vp = 6–6.5 (Meissner et al., 2004)
Vp = 6.1–6.5 (Zhao et al., 2001)
density (Haines et al., 2003)
Vp ~ 5.7 (Meissner et al., 2004)
2.5 exp (−z / 30)
2.5
2640
3010
Triassic flysch complex
2.5 exp (−z / 15)
3.0
Ordovician shallow marine strata
interbedded with andesites and volcanic tuffs
Complexly deformed early Paleozoic arcs.
Island arcs system, Precambrian gneisses,
marine sediments, volcanics and clastics.
1.2
2.3
2600
2850
2400
2.0 exp (−z / 15)
2.5
2530
2920
Beishan, UC
2.0 exp (−z / 15)
2.5
13
14
15
India LC
Greater Himalayan LC
Lhasa–Qiantang LC
0.2
0.2
0.2
2.1
2.1
2.1
2550
2880
2980
3050
3050
16
17
Qaidam–Qilian–Beishan LC
Lithospheric mantle
0.2
0.02
2.1
3.2
n
Description
1
2
Tertiary Indian foreland basin
India UC-lesser Himalayan
3
Greater Himalayan belt
4
Tethyan Himalayan sequences
5
Gangdese Batholith
6
Lhasa terrane UC
7
Qiangtang Basin, sediments
8
Qiangtang terrane UC
9
10
Songpan–Ganzi terran,
Kunlun
Qaidam Basin
11
Qaidam–Qilian Shan UC
12
Lithology
Precambrian clastic
sediments and metasediments
Late Precambrian to early Paleozoic
metasedimentary rocks tectonically overlain
by thick Permian to Cretaceous Tethyan
sedimentary sequences. Mica schist, quartzite,
paragneiss, orthogneiss, migmatite, and
Miocene leucogranite.
Unmetamorphosed to low-grade Cambrian
to Eocene shallow water shelf sediments
Cretaceous-Tertiary I-type plutonic complex
ranging from gabbro through diorite to granite
Ultramafic rocks. Ordovician and Carboniferous
to Triassic shallow marine clastic sediments.
Melange complexes
Vs = 3.4–3.8 (Mitra et al., 2005)
Vp N 5.7 (Hauck et al., 1998)
density (Haines et al., 2003)
Vp ~ 5.8–7 (Meissner et al., 2004)
Vp ~ 5.6–6.5 (Zhao et al., 2001)
Vp = 5.4–6.2 (Jiang et al., 2006)
Vp = 5–5.5 (Zhao et al., 2006)
Vp = 6–6.3 (Jiang et al., 2006)
Vp = 5.5–6.5 (Zhao et al., 2006)
Vp = 5.5–6.3 (Rui et al., 1999)
Vp = 5.9–6.2 (Rui et al., 1999)
Vs = 3.9 (Mitra et al., 2005)
Vs = 4–4.2 (Mitra et al., 2005)
density (Haines et al., 2003)
Vp = 6.6–7.1 (Meissner et al., 2004)
Vp = 7–7.3 (Zhang and Klemperer, 2005)
Vp = 6.5–7.3 (Zhao et al., 2001)
3000 Vp = 6.5–6.8 (Jiang et al., 2006)
3200[1–3.5e–5 (T–1350 °C)]
n: number of material corresponds to those from Fig. 4; H: radiogenic heat production [μW/m3]; K: thermal conductivity [W / (K ⁎ m)]; ρ: density [kg/m3]; V and
Ref.: velocities [km/s] used and references. Velocities are converted to densities using Christensen and Mooney relations (1995), the non linear relation (Table 8) and
the linear relation for the sediment layers (Table 7). From Vs to Vp we used the relation Vp/Vs = 1.73. UC: upper crust, LC: lower crust.
2006) and seismic wide angle data (Zhao et al., 2001; Meissner
et al., 2004). Haines et al. (2003) discuss the seismic reflection
and refraction data along the profile, and also propose the
density structure in two columns beneath the Lhasa and
Qiangtang terranes.
The crustal structure between the Qiangtang terrane and the
Qaidam Basin has been constructed principally from the wideangle profiles along north-eastern Tibet (Vergne et al., 2002;
Jiang et al., 2006) and from the magnetotelluric data recorded in
the PASSCAL project (Unsworth et al., 2004). These studies
suggest that the crust thins from south to north by up to 20 km
(Owens and Zandt, 1997; Li et al., 2006).
On the northernmost part of the profile, which includes the
Qaidam Basin, the Qilian Shan and Beishan Basin, we
considered the results obtained across the Golmud–Ejn transect
(Rui et al., 1999). These studies include wide-angle reflection
sounding and deep seismic reflection and refraction, magneto-
telluric, as well as gravimetry, magnetics and heat flow
measurements. The crust gets thicker again beneath the Qilian
Shan (~ 74 km) and thins to the north down to 50 km underneath
the Beishan Basin. The northern Qilian Shan reaches the higher
elevation, but the crustal root is deeper beneath the southern
Qilian Shan. Zhao et al. (2006), from seismic refraction and
wide-angle-reflection data, also provide information about the
Qaidam Basin concerning the thicknesses and compositions of
the sedimentary, upper crustal and lower crustal layers.
Finally, we used the schematic geological cross section
across the Himalayan–Tibetan orogen by Yin and Harrison
(2000) to adjust the geometry of the different crustal bodies and
their physical properties relevant to the model. Fig. 3 shows the
crustal structure inferred by all these data. The crustal thickness
in the Indian plate is between 38 and 40 km (Mitra et al., 2005).
It increases north of the MBT, reaching the maximum thickness
beneath the southern Lhasa terrane with values of 88 km
282
I. Jiménez-Munt et al. / Earth and Planetary Science Letters 267 (2008) 276–289
according to Mitra et al. (2005), or 70–80 km according to Zhao
et al. (2001). Most studies agree that the Tibetan crust is thicker
in the southern Lhasa terrane and thins to the north (Owens and
Zandt, 1997; Zhao et al., 2001; Haines et al., 2003; Li et al.,
2006). Values of the crustal thickness in the northern Lhasa
terrane are between 74 km (Owens and Zandt, 1997) and 65 km
(Haines et al., 2003), thinning northward to 60–65 km beneath
the Qiangtang terrane (Owens and Zandt, 1997; Zhao et al.,
2001; Haines et al., 2003; Li et al., 2006), and to 55–60 km in
the Songpan–Ganzi terrane and Qaidam Basin (Owens and
Zandt, 1997; Zhao et al., 2001; Kind et al., 2002; Li et al.,
2006). The Moho step at the Kunlun–Qaidam border is
associated with a change in topography. The several bodies
from Fig. 3 mainly correspond to the different terranes forming
the plateau where we have differentiated sediments, upper and
lower crust. The values of the physical properties considered for
each body are listed in Table 1. Thermal conductivities and
radiogenic heat production for the lithologies described from each
Fig. 4. Modelling results. The top three graphs (a–c) show data and results predicted from the model: a) Bouguer gravity anomaly, b) geoid and c) elevation. The data
are represented by dots with bars, corresponding to the standard deviation within a range of 50 km to each side of the profile. Continuous blue lines are the modelled
values. d) Lithospheric model with a vertical exaggeration of 2.4. The dots show earthquake hypocenters (Engdahl et al., 1998) projected onto the model from a strip
150 km wide each side of the model.
I. Jiménez-Munt et al. / Earth and Planetary Science Letters 267 (2008) 276–289
terrane have been taken from the literature, and densities have also
been taken from literature or calculated from Vp-density
relationships when possible (Christensen and Mooney, 1995).
4. Results
An initial consideration of the data, the topography and the
compiled crustal structure, suggests some influence of the deep
lithosphere on the isostasy of the Tibetan Plateau. The nearly
uniform elevation of the plateau in spite of significant variations
in crustal thickness from north to south (Fig. 3) requires corresponding variations in the upper mantle to maintain isostatic
compensation. The observed thickness of the crust beneath the
Himalaya and southern Tibet is enough to provide the high
measured elevation; in contrast, the thinner crust in the central
and northern plateau should result in a decrease of about 1400 m
in elevation if a constant lithosphere thickness is considered.
The relatively abundant available data allowed us to
reasonably fix the initial crustal structure and properties along
the profile. Further changes in crustal thickness and densities
were slight and restricted to the uncertainty range reported by
the different authors and our studies. Consequently, we changed
the depth of the lithosphere to fit the measured elevation,
gravity and geoid.
Fig. 4 shows the observed and calculated values of Bouguer
anomaly (Fig. 4a), geoid (Fig. 4b) and elevation (Fig. 4c) for the
283
lithosphere structure (Fig. 4d) that best fits all these observables.
The data are represented by dots with dispersion bars, corresponding to the standard deviation within a range of 50 km to
each side of the profile, and the continuous lines are the calculated values from the assumed model (Fig. 4d). In general, the
model is in good agreement with the observables with the
exception of some local misfits.
The conspicuous variations of the depth of the lithosphere–
asthenosphere boundary are the most remarkable feature. The
base of the lithosphere varies from 148 km beneath the Indian
plate to the maximum of 264 km beneath the southern Tibetan
Plateau. Northwards it thins abruptly to 98 km under the
northern Lhasa and Qiantang terranes, over a region of 400 km
width. The decreasing topography north of the Kunlun fault is
associated with a deeper base of the lithosphere, with the
maximum of 177 km underneath the northern Qaidam Basin
and southern Qilian Shan. This lithosphere structure implies
hotter lithospheric mantle and lower crust beneath the northern
Tibetan Plateau. These results are in agreement with those
inferred by several seismic studies (Haines et al., 2003; Tilmann
and Ni, 2003) and review papers (DeCelles et al., 2002) where
an asthenospheric upwelling is suggested underneath the
northern Tibetan Plateau.
Fig. 4d also shows the earthquake hypocenters within a
band 150 km on each side of the profile (Engdahl et al., 1998).
Note that the deeper earthquakes occur south of the MCT, affecting
Fig. 5. a) Surface heat flow: predicted from the model (blue line) and, measurement (red dots) from Rui et al. (1999) along the profile between Qaidam and Beishan
basin and global data set compiled by Pollack et al. (1993). Heat flow measurements have been projected from a 200 km half-width strip on to the profile, with the error
bar dependent with the data quality (A = 0.10%, B = 0.15%, C = 0.20%, D = 0.25%); b) lithosphere temperature distribution predicted from the model (vertical
exaggeration of 3). The dots show earthquake hypocenters (Engdahl et al., 1998) projected onto the model from a strip 150 km wide each side of the model. Thick
horizontal bars are temperatures obtained from different studies, with the vertical lines indicating the range of values: seismic detection of the α-β quartz transition
(Mechie et al., 2004, [A]), analyses on xenolith samples (Hacker et al., 2000, [B]) and, determined temperatures from P-wave velocity data and an assumed
composition (Galve et al., 2006, [C]) (see Fig. 6).
284
I. Jiménez-Munt et al. / Earth and Planetary Science Letters 267 (2008) 276–289
the highly deformed lower crust and the uppermost mantle beneath
the Greater Himalayan belt where the lithosphere is thicker and
colder. Beneath the north Lhasa block and Qiangtang all the
seismicity occurs within the upper and middle crust.
Fig. 5 shows the modelled surface heat flow and the lithosphere temperature distribution. The Indian plate is characterized by horizontal isotherms with moderate temperatures within
the crust and the lithosphere mantle, and a Moho temperature of
about 550 °C. The crustal thickening beneath the Greater
Himalaya deflects the isotherms upwards near the crust-mantle
boundary due to thermal refraction produced by the lower
thermal conductivity of the crust relative to the lithospheric
mantle. Moho temperature in this region ranges between 900
and 950 °C increasing northwards to 1150 °C. The Qiangtang
region displays the maximum temperatures due to pronounced
lithosphere thinning with Moho temperature ~ 1200 °C. Farther
to the north, beneath the Qaidam, Qilian and Beishan regions,
crustal temperatures decrease again showing large variations at
the Moho ranging from 900 °C to less than 700 °C in Beishan.
The upper panel (Fig. 5a) shows the measured surface heat flow
available (Pollack et al., 1993; Rui et al., 1999) projected from a
200 km half-width strip on to the profile and the heat flow
predicted by our model. Our results show large variations from
~ 50 mW/m2 in the Indian plate and the Beishan region, to
maximum values of 85–90 mW/m2 in the northern Lhasa and
Qiangtang regions. Despite the large scattered and uncertainties
associated with the reported heat flow measurements, the
general trend of measured and calculated values coincide pretty
well showing low values in India, the highest in Himalaya and
Lhasa block, and minima in Qaidam basin. Moreover, the
calculated values at the southern and northern ends of the profile
match those measured in stable lithosphere, and those in the
northern Lhasa and Qiangtang regions compare well with
young deformed regions elsewhere.
Fig. 6 compares the temperature and pressure/depth
predicted by our model and those P–T conditions derived
from seismic data, xenoliths and the α-β quartz transition for
several lithosphere columns along the modelled transect. Galve
et al. (2006) suggested temperatures from P-wave velocities
beneath the Tethyan Himalayan of ~ 500 to 850 °C between 1
and 2 GPa. The predicted geotherm for our model in the lower
crust of the Greater Himalaya is between the extremes suggested by Galve et al. (2006). However, beneath the Qiangtang
and northern Lhasa block, higher temperatures compared with
those from southern plateau are observed (Fig. 6) (Hacker et al.,
2000; Mechie et al., 2004; Galve et al., 2006). Using a radiogenic heat production distribution in the Qiangtang and Lhasa
terranes similar to that assumed for the Songpan–Ganzi and the
Himalayan belt, the resulting Qiangtang temperatures at 0.8 and
1.3 GPa are ~ 600 and 800 °C respectively, and a bit lower
beneath north Lhasa block. These values fall within the
temperature range proposed by Galve et al., (2006), but they
are far away from those inferred from the α-β quartz transition
(700–800 °C at 0.5–0.85 GPa, Mechie et al., 2004) and from
analysis of xenoliths (800–900 °C at 0.8–1.3 GPa, Hacker
et al., 2000). This indicates that the pronounced lithospheric
thinning affecting the Lhasa and Qiangtang regions does not
Fig. 6. Temperature in depth resulting from the model compared with some
external data. Colour pattern shows the lower crustal temperatures predicted by
seismic velocities (Galve et al., 2006, [C]). Yellow ellipse shows the predicted
temperatures in southern Qiangtang and northern Lhasa block from the seismic
detection of the α-β quartz transition (Mechie et al., 2004, [A]). Pink ellipse
shows temperatures obtained by Hacker et al. (2000) [B] in central Qiangtang
from xenoliths. Black dashed lines are granite solidus curves for (a) dry melting,
(b) dehydration melting, and (c) wet melting. Red, green and blue lines are the
resulting geotherms of our best fitting model beneath Himalaya, north Lhasa and
Qiangtang terranes, respectively. Green and blue dashed lines are the geotherms
of a model with lower heat production on the Lhasa and Qiangtang terranes.
suffice to explain such high temperatures at middle crustal
levels. Moreover, some reflections with anomalous amplitudes
(bright spots) observed at depths of 15 to 18 km on seismic lines
have been interpreted as granitic magmas derived by partial
melting of the tectonically thickened crust (Brown et al., 1996),
thus supporting high temperatures at upper crustal levels.
In order to match these high crustal temperatures we have
considered a higher characteristic length scale in the exponential
radiogenic heat distribution. Therefore, we used a value of
30 km for the Qiangtang and Lhasa terranes instead of the
common value of 15 km used in the upper to middle crust of the
other areas (Table 1). In contrast to the other domains, the thick
Tibetan crust is not formed from a cratonic terrane where
the heat producing elements are mainly concentrated in the
upper crust through magmatic and metamorphic processes that
occurred at older times, but it has been formed by young
tectonic processes that have piled up crustal slices and doubled
the thickness of the crust (Dewey et al., 1988). This suggests
that we should consider a smaller downward decay of the
radiogenic heat production in the upper and middle crust of the
Lhasa and Qiangtang terranes. This smaller decay of the
radiogenic heat production on the plateau was also required by
Jiménez-Munt and Platt (2006) in their modelling of the uplift
I. Jiménez-Munt et al. / Earth and Planetary Science Letters 267 (2008) 276–289
history of the plateau. Using this smaller value, Fig. 6 (solid
lines) shows a good fit between the modelled geotherms
beneath the Lhasa and Qiangtang terranes and those inferred
from seismic, xenolith and α-β quartz transition data, although
the lower crustal temperatures are just on the upper limit
proposed by Galve et al. (2006). It is worth noting that the
predicted Vp values (Galve et al., 2006) vary in each terrane
(e.g. from 6.4 to 6.6 km/s on the Greater Himalaya) and,
therefore, the P–T domains could be slightly larger than those
presented on the figure. The model predicts Moho temperatures
beneath the Himalaya of ~ 925 °C, whereas higher temperatures
of ~ 1180 °C are obtained beneath the northern Lhasa and
Qiangtang terranes.
Our transect crosses the Bayan Har domain much more
further to the west than Galvé's profile (Galve et al., 2006),
where this domain becomes narrowest (~ 100 km). Our calculated temperatures in the Bayan Har are lower than those from
Qiangtang terrane, and higher than those proposed by Galve
et al. (2006) who assumes a felsic composition (granite and
granodiorite). However, Vergne et al. (2002) propose a more
mafic/ultramafic rock composition according to the high Vp/Vs
reported in the ~ 50 km-wide strip south of Kunlun Fault. In this
case, the temperatures derived from (Galve et al., 2006) would
increase noticeably and our results would fit better.
5. Discussion
The modelling approach used in this study integrates
different observables: topography, Bouguer anomaly, geoid
height, and temperature within the crust. These variables are
governed by the density and temperature distribution within the
lithosphere, which are constrained by the available geological
and seismic data. The main assumptions of the model are that
the lithosphere is in thermal steady state and is supported by
local isostasy. These assumptions impose some limitations in
interpreting the resulting lithosphere structure.
Local isostasy is an acceptable approximation, as gravity
anomalies show that over length scales of a few hundreds
kilometres, density contrasts within the lithosphere are isostatically compensated (England and Molnar, 1997). Furthermore, the gravity field in Tibet indicates that the topography
there is for the most part in local isostatic compensation (Jin
et al., 1994).
The assumption of thermal steady state is valid mostly in old
lithospheres. The Himalaya and Tibetan Plateau are regions of
tectonic activity that has been rapid on the time-scale for thermal
equilibration of the lithosphere, and this activity is likely to have
produced transient perturbations in the temperature distribution.
However, quantifying these temperature variations requires a
much better knowledge of the geodynamic evolution of the
region. We opted for the steady-state calculations since the timing
of thickness variations is too uncertain to calculate non-steadystate temperature distributions. On the other hand, this model
accounts for a lithospheric density-depth distribution that is
compatible with elevation, gravity, and geoid observations
independent of the actual temperature distribution. The calculation in steady state minimizes the variations of lithospheric
285
thickness, since the modelling tends to underestimate the
lithosphere thickness when thickening processes occur under
transient conditions, and overestimate it for thinning processes.
Therefore, the lithosphere–asthenosphere boundary beneath the
Himalayan and southern Tibetan Plateau could be even deeper,
whereas underneath the north Lhasa block and Qiangtang the
lithosphere could be thinner. Although the numerical code also
permits a temperature dependent crustal density, we have not
linked the density of crustal bodies with temperature. Instead, we
have derived crustal densities from measured Vp and Vs values
which already incorporate compositional changes and P–T
variations.
The practically constant lithospheric thickness of the Indian
plate (145 km) abruptly increases north of the Main Boundary
Thrust, reaching a maximum of 260 km beneath the south Lhasa
block. This result agrees with the model proposed by Jin et al.
(1996) where the dense mantle portion of the Indian plate
has been subducted beneath the Eurasian plate for a distance
of 500–700 km north of the Main Boundary Thrust. This
boundary of the underthrusting Indian lithosphere is also
delimited by the faster P-wave velocities in the south and slower
velocities in the north of the plateau (Hearn et al., 2004).
The nearly uniform topography of the Tibetan Plateau masks
the large variations in the crustal and mantle structure between
the southern and northern parts. Our results show that the
characteristics at depth of the northern part of the Tibetan
Plateau are completely different from the south, and thinner
lithospheric mantle and higher middle and lower crustal temperatures are the most remarkable features. These north-south
variations are also indicated by seismic tomography data;
relatively slow Pn wave velocity and inefficient Sn propagation
beneath the northern half of the plateau, but relatively fast
Pn wave velocity and efficient Sn propagation beneath the
southern half (McNamara et al., 1997; Owens and Zandt, 1997;
Tapponnier et al., 2001). Further evidences supporting a higher
thermal regime in northern Tibet have long been suggested by
several studies. From receiver functions, Kind et al. (2002)
found that although the 410 and 660 km mantle discontinuities
are parallel, both are depressed by about 20 km beneath the
northern plateau relative to the south. They explain that by a
northward increase in the average temperature of about 300 °C
between the surface and 410 km depth. Rapine et al. (2003)
find low lower-crustal S-wave velocities and the absence of an
S-wave velocity gradient in the mantle beneath northern Tibet
from dispersion studies, consistent with a hot and weak upper
mantle. A recent global thermal model (Artemieva, 2006) also
predicts higher temperatures beneath north-east Tibet than in the
south. The temperatures calculated by Artemieva (2006) at
50 km depth vary from 600 °C in the south Tibet to 700 °C in
the NE Tibet. At deeper levels (150 km), calculated temperatures vary from 1100 °C to 1300 °C from S to NE Tibet. The
estimated lithospheric thickness, according to Artemieva
(2006), varies from 180 km beneath S Tibet to 125 km beneath
NE Tibet. Basaltic volcanism of different ages, but continuing
until the most recent past (less than 5 Ma), is known to have
occurred in northern Tibet (Yin and Harrison, 2000). Quaternary volcanism (Turner et al., 1996; Tapponnier et al., 2001) and
286
I. Jiménez-Munt et al. / Earth and Planetary Science Letters 267 (2008) 276–289
the unusually low P-velocities in the lower crust and upper
mantle from wide-angle and refraction studies (Meissner et al.,
2004), and Pn tomography (McNamara et al., 1997; Haines
et al., 2003) fit perfectly into the concept of a hot, attenuated and
weak lithosphere. The warm and weak layers in the lower crust
and the upper mantle in the Qiangtang appear to be facilitating
an easterly escape of the whole lithosphere; this eastward
movement is observed by GPS data (Zhang et al., 2004) and
also suggested by the strong east–west anisotropy (Meissner
et al., 2004).
Haines et al. (2003) conclude from a crustal velocity model
around the Bangong–Nujiang Suture (BNS) that the total
pressure beneath the Lhasa and Qiangtang blocks at 65 km is
the same (within 3 MPa), so that large mantle density differences to great depth below the BNS are precluded. Our
results are in agreement with the lack of density variations south
and north around the BNS, but large variations are predicted for
the northern and southern Lhasa terrane. The good fit of
topography, geoid, and gravity data to our model (Fig. 4), shows
that crustal isostatic compensation of the plateau is insufficient
to explain observations and requires density variations to be
present in the deeper lithosphere.
Our predicted lithospheric thickness also agrees with the
pattern of the elastic thickness of the lithosphere (Te) obtained
by a non-spectral technique from observed topography and
Bouguer gravity anomaly data (Jordan and Watts, 2005). Considering that for temperatures higher than ~ 800 °C the strength
of the lithosphere is practically zero, we can compare the
thickness of the mechanical lithosphere with the depth of this
isotherm. From our modelling we find that the deepest regions
of the 800 °C isotherm are underneath the Himalayan foreland
(~ 72 km), Qaidam Basin (~ 58 km), and in Beishan (~ 61 km).
In contrast, the 800 °C isotherm is at about 30 km depth beneath
the northern Lhasa and Qiangtang terranes. This corresponds
with the results of Jordan and Watts (2005), who found that the
Himalayan foreland forms a rigid block with a thick mechanical
lithosphere or equivalent elastic thickness (Te = 40–100 km),
together with the Qaidam Basin to the north and Sichuan Basin
to the east; and that the Tibetan Plateau is a weaker region
(Te = 0–20 km) that separates these rigid blocks. Moreover, we
found that the average temperature of the lithospheric mantle
beneath the Qaidam Basin is less than that of the Qilian Shan.
This suggests that the Qaidam Basin has a higher lithospheric
strength than the Qilian Shan, which could explain the
transmission of stress through to the north of the Qaidam
Basin, causing the ongoing deformation in the Qilian Shan.
5.1. Geodynamic implications
Many geodynamic models of the Tibetan Plateau invoke
continuum deformation of the lithosphere, with shortening and
thickening of both crust and lithospheric mantle (e.g., Houseman and England, 1993). Some of these models suggest that the
thickened Tibetan lithosphere become unstable, and that the
lower part was delaminated or otherwise removed, producing a
localized upwelling of the asthenosphere (Molnar et al., 1993;
Jiménez-Munt and Platt, 2006 and references therein). Establishing the geometry of the crustal and lithospheric structures is
a key to understand the evolution of the Himalayan–Tibetan
orogen.
Recent thin sheet numerical modeling indicates that rapid
removal of the lithospheric root beneath the thickened crust of
Tibet successfully explains the current elevation of the plateau,
its lack of surface slope, the steep south and north margins, and
the pattern of the present deformation (Jiménez-Munt and Platt,
2006), using the methodology describe in Jiménez-Munt et al.
(2005). Two important additional results can be extracted. First,
their modeling suggests that this removal took place within the
last 12 My. The second result refers to the sharp boundary along
the northern topographic edge of the Tibetan Plateau. This
boundary can only be produced, subsequent to lithospheric
thinning, by the continuous thickening of both crust and
lithospheric mantle beneath the northern plateau. Thickening of
the lithospheric mantle is the combined product of both
Fig. 7. Cartoon of the resulting profile, lithosphere scale.
I. Jiménez-Munt et al. / Earth and Planetary Science Letters 267 (2008) 276–289
shortening and thermal relaxation. A similar evolution can be
inferred for the Tethyan Himalaya and southern Tibetan Plateau.
There, continuous convergence between India and Asia has
thickened the crust as well as the upper mantle, and thus modified the topography produced during lithospheric thinning at
about 10 Ma.
The N–S variations of the characteristics of the lithosphere
suggest some consequences for the evolution of the Tibetan
Plateau. The thin and hot lithospheric mantle beneath the
northern Lhasa block and Qiangtang (Figs. 4 and 5) with the
onset of strong shear wave splitting anisotropy about 25 km
south of the BNS, indicate that part of the thickened lithospheric
mantle has been removed (Fig. 7), resulting in an upwelling of
asthenospheric mantle (Molnar et al., 1993; Jiménez-Munt and
Platt, 2006). Elevated temperatures and perhaps partial melting
in the upper mantle would be a natural consequence of this
process, and would produce the low electrical resistivities
(Partzsch et al., 2000). Moreover, as mantle-derived melts
migrate upward through the crust, they initiate crustal melting
and pervasive low magnetic resistivity results in the lower crust
(Unsworth et al., 2004). Our results are consistent with a model
in which Asian lithosphere extends as far south as the Kunlun
Fault and the upper mantle beneath the Songpan–Ganze and
Qiangtang terranes is sufficiently hot to contain a small fraction
of interconnected partial melt. The crust thins from south to
north with a concomitant increase in Poisson's ratio from normal values in the south to unusually high values in the north,
which also suggests that the crust of the northern plateau is
partially melted due to high temperatures (Owens and Zandt,
1997). All this is also in agreement with the observed Quaternary
volcanism (Turner et al., 1996; Tapponnier et al., 2001).
The lithospheric structure we obtain could therefore be in
agreement with removal of the lithospheric root beneath the
thickened crust of Tibet at ~ 8 My (Molnar et al., 1993) or within
the last 12 My (Jiménez-Munt and Platt, 2006). Our model
makes good sense in terms of the post-12 Ma convergence
history, since convergence at 20 mm/yr for 12 My gives 240 km
underthrusting, and we can be sure that Indian lithosphere
extended at least as far as the Indus–Tsangpo suture before then.
Moreover, our proposed lithospheric geometry is in agreement
with those works that propose bidirectional underthrusting from
north and south beneath Tibet. Yin and Harrison (2000) argue
that the occurrence of syn-collisional calc-alkline type volcanism in southern and central Tibet appears to require that some
portion of the continental crust from both north and south must
have been subducted into the mantle beneath Tibet. Rui et al.
(1999) also suggest from deep seismic reflection profiles along
the Northern Qilian Shan that the Ala Shan underthrusts the
northern Tibetan Plateau.
6. Conclusions
Joint modelling of gravimetry, topography and geoid data,
using geological and crustal seismic and thermal data as
constraints, allowed us to establish a new model of the lithospheric structure of the Tibetan Plateau and its surroundings.
This study allows us to image the variations of the lithosphere
287
thickness along a profile from the Indian plate to Beishan
Basin in China, crossing the Himalayan belt and the Tibetan
Plateau.
These results suggest that the coldest part of the mantle
underlies southern Tibet and that the hottest underlies north
central Tibet. The present average topography of the High
Himalaya and Tibetan Plateau is supported by a non-uniform
crustal and lithospheric structure. The High Himalaya and the
southern Tibetan Plateau consist of thick crust and thick upper
mantle whereas the northern Tibetan Plateau is underlain by a
thinner crust and thin lithospheric mantle. The surface manifestations of such variations in mantle structure are obscured by the
nearly uniform elevation of Tibet, the height of which is
compensated by thick crust in the south and by hot upper mantle
to the north. One of the principal conclusions is that the Tibetan
Plateau as a whole cannot be supported isostatically only by a
thickened crust. A thin and hot lithosphere beneath the northern
plateau is required to explain the high topography, and the
measured gravity and geoid.
The crust beneath southern Tibet is thicker than 70 km, and
presumably some of this crust is Indian. Therefore, the Indian
lithosphere appears to be being emplaced directly beneath the
southern Tibetan crust. This suggests removal of Tibetan lithospheric mantle before (or during) the underthrusting of India.
This corresponds well with geodynamic ideas concerning the
evolution of the Indian plate (underthrusting or subduction?)
and the removal of lithosphere, with the resulting upwelling of
asthenosphere.
The nearly uniform elevation of the plateau above a laterally
varying upper mantle is associated with corresponding variations in crustal thickness to maintain isostatic compensation.
Therefore, there are different modes of isostatic support of the
plateau, the southern part (Tethyan Himalaya and southern
Lhasa block) by thick-low density crust and the northern part
(northern Lhasa block and Qiangtang) by hot, low-density upper
mantle.
We propose in this paper that lithospheric thinning was
the consequence of rapid large-scale removal lithosphere
at some time between 8 and 12 Ma, in agreement with
geological observations as well as numerical modelling results
(e.g., Molnar et al., 1993; Turner et al., 1996; Jiménez-Munt
and Platt, 2006). According to the modeling by Jiménez-Munt
and Platt (2006), the sudden replacement of cold mantle
lithosphere by hot asthenosphere produced a rapid increase of
the high Himalaya and Tibetan Plateau by about 1.5 km. The
resulting plateau still sloped significantly northwards, however, reflecting the previous pattern of crustal thickening. The
final increase in elevation of the northern part of the plateau,
producing the present uniform elevation and the abrupt
northern margin, took place about 10 My after the replacement
of the upper mantle, and was the product of south to north
crustal flow, continued shortening in the north, and lithospheric thermal recovery (Jiménez-Munt and Platt, 2006).
Continued convergence between the Indian and Asian plates is
also moving Indian lithosphere beneath the Tibetan Plateau,
thickening the crust and replacing asthenospheric mantle with
Indian lithosphere.
288
I. Jiménez-Munt et al. / Earth and Planetary Science Letters 267 (2008) 276–289
Acknowledgements
This is a contribution of the Team Consolider-Ingenio 2010
(CSD2006-00041). This research was also partly financed
by Projects WESTMED (REN 2002-11230-E-MAR/01-LECEMA22F), SAGAS (CTM2005-08071-C03-03/MAR) and
TopoAtlas (CGL2006-05493/BTE). I. Jiménez-Munt was
supported by the Spanish Government by the Program
‘Ramón y Cajal’. The authors thank J. Fullea for developing
the program used to calculate the Bouguer anomaly and
A. Galvé for give us her figure used in Fig. 6. Constructive
reviews from anonymous reviewer helped us improve the
final version.
References
Alsdorf, D., Nelson, K.D., 1999. Tibetan satellite magnetic low: evidence for
widespread melt in the Tibetan crust? Geology 27, 943–947.
Artemieva, I.M., 2006. Global 1° × 1° thermal model TC1 for the continental
lithosphere: implications for lithosphere secular evolution. Tectonophysics
416, 245–277.
Braitenberg, C., Zadro, M., Fang, J., Wang, Y., Hsu, H.T., 2000. The gravity and
isostatic Moho undulations in Qinghai–Tibet plateau. J. Geodyn. 30,
489–505.
Brown, L.D., Zhao, W., Nelson, K.D., Hauck, M., Alsdorf, D., Ross, A., Cogan,
M., Clark, M., Liu, X., Che, J., 1996. Bright spots, structure, and magmatism
in Southern Tibet from INDEPTH seismic reflection profiling. Science 274,
1688–1690.
Christensen, N.I., Mooney, W.D., 1995. Seismic velocity structure and
composition of the continental crust: a global view. J. Geophys. Res. 100,
9761–9788.
Chung, S.-L., Chu, M.-F., Zhang, Y., Xie, Y., Lo, C.-H., Lee, T.-Y., Lan, C.-Y., Li, X.,
Zhang, Q., Wang, Y., 2005. Tibetan tectonic evolution inferred from spatial and
temporal variations in post-collisional magmatism. Earth-Science Rev. 68,
173–196.
DeCelles, P.G., Robinson, D.M., Zandt, G., 2002. Implications of shortening in
the Himalayan fold-thrust belt for uplift of the Tibetan Plateau. Tectonics 21,
1062. doi:10.1029/2001TC001322.
Dewey, J.F., Shackelton, R.M., Chang, C., Sun, Y., 1988. The tectonic evolution
of the Tibetan Plateau. Phil. Trans. R. Soc. Lond. A327, 379–413.
Engdahl, E.R., van der Hilst, R., Buland, R., 1998. Global teleseismic
earthquake relocation with improved travel times and procedures for depth
determination. Bull. Seismol. Soc. Am. 88, 722–743.
England, P.C., Houseman, G.A., 1989. Extension during continental convergence,
with application to the Tibetan plateau. J. Geophys. Res. 94, 17,561–17,579.
England, P.C., Molnar, P., 1997. Active deformation of Asia: from kinematics to
dynamics. Science 278, 647–650.
Fernàndez, M., Marzán, I., Torne, M., 2004a. Lithospheric transition from the
Variscan Iberian Massif to the Jurassic oceanic crust of the Central Atlantic.
Tectonophysics 386, 97–115.
Fernàndez, M., Torne, M., Garcia-Castellanos, D., Vergés, J., Wheeler, W.,
Karpuz, R., 2004b. Deep structure of the Vøring Margin: the transition from
a continental shield to a young oceanic lithosphere. Earth Planet. Sci. Lett.
221, 131–144.
Fullea, J., Fernandez, M., Zeyen, H., submitted for publication. FA2BOUG- a
FORTRAN 90 code to compute Bouguer gravity anomalies from gridded
free air anomalies: application to the Atlantic-Mediterranean transition zone,
submitted to Computers and Geosciences.
Galve, A., Jiang, M., Hirn, A., Sapin, M., Laigle, M., de Voogd, B., Gallart, J.,
Qian, H., 2006. Explosion seismic P and S velocity and attenuation constraints
on the lower crust of the North–Central Tibetan Plateau, and comparison with
the Tethyan Himalayas: Implications on composition, mineralogy, temperature, and tectonic evolution. Tectonophysics 412, 141–157.
Galvé, A., Sapin, M., Hirn, A., Diaz, J., Lepine, J.-C., Laigle, M., Gallart, J., Jiang,
M., 2002. Complex images of Moho and variation of Vp/Vs across the Himalaya
and South Tibet, from a joint receiver-function and wide-angle-reflection
approach. Geophys. Res. Lett. 29, 2182. doi:10.1029/2002GL015611.
Hacker, B.R., Gnos, E., Ratschbacher, L., Grove, M., McWilliams, M., Sobolev,
S.V., Wan, J., Zhenhan, W., 2000. Hot and dry deep crustal xenoliths from
Tibet. Science 287, 2463–2466.
Haines, S.S., Klemperer, S.L., Brown, L., Jingru, G., Mechie, J., Meissner, R.,
Ross, A., Wenjin, Z., 2003. INDEPTH III seismic data: From surface
observations to deep crustal processes in Tibet. Tectonics 22, 1001.
doi:10.1029/2001TC001305.
Hauck, M.L., Nelson, K.D., Brown, L.D., Zhao, W., Ross, A.R., 1998. Crustal
structure of the Himalayan orogen at ~90° east longitude from Project
INDEPTH deep reflection profiles. Tectonics 17, 481–500.
Hearn, T.M., Wang, S., Ni, J.F., Xu, Z., Yu, Y., Zhang, X., 2004. Uppermost
mantle velocities beneath China and surrounding regions. J. Geophys.
Res. 109. doi:10.1029/2003JB002874.
Hirn, A., et al., 1995. Seismic anisotropy as an indicator of mantle flow beneath
the Himalayas and Tibet. Nature 375, 571–574.
Houseman, G., England, P., 1993. Crustal thickening versus lateral expulsion in
the Indian-Asian continental collision. J. Geophys. Res. 98, 12233–12249.
Jiang, M., Galvé, A., Hirn, A., de Voogd, B., Laigle, M., Su, H.P., Diaz, J., Lepine,
J.C., Wang, Y.X., 2006. Crustal thickening and variations in architecture from
the Qaidam Basin to the Qang Tang (North-Central Tibetan Plateau) from
wide-angle reflection seismology. Tectonophysics 412, 121–140.
Jiménez-Munt, I., Garcia-Castellanos, D., Fernandez, M., 2005. Thin-sheet
modelling of lithospheric deformation and surface mass transport.
Tectonophysics 407, 239–255.
Jiménez-Munt, I., Platt, J.P., 2006. Influence of mantle dynamics on the
topographic evolution of the Tibetan Plateau: results from numerical
modelling. Tectonics 25, TC6002. doi:10.1029/2006TC001963.
Jin, Y., McNutt, M.K., Zhu, Y., 1994. Evidence from gravity and topography
data for folding of Tibet. Nature 371, 669–674.
Jin, Y., McNutt, M.K., Zhu, Y.-S., 1996. Mapping the descent of Indian and
Eurasian plates beneath the Tibetan Plateau from gravity anomalies.
J. Geophys. Res. 101, 11275–11290.
Jordan, T.A., Watts, A.B., 2005. Gravity anomalies, flexure and the elastic
thickness structure of the India–Eurasia collisional system. Earth Planet.
Sci. Lett. 236, 732–750.
Kind, R., Yuan, X., Saul, J., Nelson, D., Sobolev, S.V., Mechie, J., Zhao, W.,
Kosarev, G., Ni, J., Achauer, U., Jiang, M., 2002. Seismic images of crust
and upper mantle beneath Tibet: evidence for Eurasian plate subduction.
Science 298, 1219–1221.
Lachenbruch, A., Morgan, P., 1990. Continental extension, magmatism and
elevation; formal relations and rules of thumb. Tectonophysics 174, 39–62.
Lemoine, F.G., et al., 1998. The development of the Joint NASA GSFC and
NIMA geopotential model EGM96. NASA Goddard Space Flight Center,
Greenbelt, Md.
Li, S., Mooney, W.D., Fan, J., 2006. Crustal structure of mainland China from
deep seismic sounding data. Tectonophysics 420, 239–252.
McNamara, D.E., Walter, W.R., Owens, T.J., Ammon, C.J., 1997. Upper mantle
velocity structure beneath the Tibetan Plateau from Pn travel time
tomography. J. Geophys. Res. 102, 493–505.
Mechie, J., Sobolev, S.V., Ratschbacher, L., Babeyko, A.Y., Bock, G., Jones, A.G.,
Nelson, K.D., Solon, K.D., Brown, L.D., Zhao, W., 2004. Precise temperature
estimation in the Tibetan crust from seismic detection of the α-β quartz
transition. Geology 32, 601–604.
Meissner, R., Tilmann, F., Haines, S., 2004. About the lithospheric structure of
central Tibet, based on seismic data from the INDEPTH III profile.
Tectonophysics 380, 1–25.
Mitra, S., Priestley, K., Bhattacharyya, A.K., Gaur, V.K., 2005. Crustal structure
and earthquake focal depths beneath northeastern India and southern Tibet.
Geophys. J. Int. 160, 227–248.
Molnar, P., England, P., Martinod, J., 1993. Mantle dynamics, uplift of the
Tibetan Plateau and the Indian monsoon. Rev. Geophys. 31, 357–396.
Molnar, P., Lyon-Caen, H., 1989. Fault plane solutions of earthquakes and active
tectonics of the northern and eastern parts of the Tibetan Plateau. Geophys.
J. Int. 99, 123–153.
I. Jiménez-Munt et al. / Earth and Planetary Science Letters 267 (2008) 276–289
Murphy, M.A., Yin, A., Garrison, T.M., Durr, S.B., Chen, Z., et al., 1997.
Significant crustal shortening in southcentral Tibet prior to the Indo–Asian
collision. Geology 25, 719–722.
Naqvi, S., Rogers, J., 1987. The Precambrian Geology of India. Oxford
University Press Inc, USA.
Nelson, K.D., Zhao, W., Brown, L.D., Kuo, J., Che, J., Liu, X., Klemperer, S.L.,
Makovsky, Y., Meissner, R., Mechie, J., Kind, R., Wenzel, F., Ni, J.,
Nabelek, J., Leshou, C., Tan, H., Wei, W., Jones, A.G., Booker, J.,
Unsworth, M., Kidd, W.S.F., Hauck, M., Alsdorf, D., Ross, A., Cogan, M.,
Wu, C., Sandvol, E., Edwards, M., 1996. Partially molten middle crust
beneath southern Tibet: synthesis of project INDEPTH results. Science 274,
1684–1688.
Owens, T.J., Zandt, G., 1997. Implications of crustal property variations for
models of Tibetan plateau evolution. Nature 387, 37–43.
Partzsch, G.M., Schilling, F.R., Arndt, J., 2000. The influence of partial melting
on the electrical behavior of crustal rocks: laboratory examinations, model
calculations and geological interpretations. Tectonophysics 317, 189–203.
Platt, J.P., England, P., 1994. Convective removal of lithosphere beneath mountain
belts: thermal and mechanical consequences. Am. J. Sci. 294, 307–336.
Pollack, H.N., Hurter, S.J., Johnson, J.R., 1993. Heat flow from the Earth's
interior: analysis of the global data set. Rev. Geophys. 31, 267–280.
Rapine, R., Tilmann, F., West, M., Ni, J., Rodgers, A., 2003. Crustal structure of
northern and southern Tibet from surface wave dispersion analysis.
J. Geophys. Res. 108. doi:10.1029/2001JB000445.
Rui, G., Xiangzhou, C., Gongjian, W., 1999. Lithospheric structure and
geodynamic model of the Golmud–Ejn transect in northern Tibet. Geol. Soc.
Am. 328, 9–17.
Sandwell, D.T., Smith, W.H.F., 1997. Marine gravity from Geosat and ERS 1
satellite altimetry. J. Geophys. Res. 102, 10,039–10,054.
Talwani, M., Worzel, J.L., Landisman, M., 1959. Rapid gravity computations
for two-dimensional bodies with application to the Mendocino submarine
fracture zone. J. Geophys. Res. 64, 49–59.
Tapponnier, P., Zhiqin, X., Roger, F., Meyer, B., Arnaud, N., Wittlinger, G.,
Jingsui, Y., 2001. Oblique stepwise rise and growth of the Tibetan Plateau.
Science, 294, 1671–1677.
Tilmann, F., Ni, J., 2003. INDEPTH III seismic team, seismic imaging of the
downwelling Indian lithosphere beneath central Tibet. Science 300,
1424–1427.
Turner, S., Arnaud, N., Liu, J., Rogers, N., Hawkesworth, C., Harris, N., Kelley,
S., van Calsteren, P., Deng, W.M., 1996. Postcollisional, shoshonitic
volcanism on the Tibetan plateau: implications for convective thinning of the
lithosphere and the source of ocean island basalts. J. Petrol. 37, 45–71.
289
Turner, S., Hawkesworth, C.J., Liu, L., Rogers, N., Kelley, S., van Calsteren, P.,
1993. Timing of Tibetan uplift constrained by analysis of volcanic rocks.
Nature 364, 50–53.
Unsworth, M., Wei, W., Jones, A.G., Li, S., Bedrosian, P., Booker, J., Jin, S.,
Deng, M., Tan, H., 2004. Crustal and upper mantle structure of northern
Tibet imaged with magnetotelluric data. J. Geophys. Res. 109, B02403.
doi:10.1029/2002JB002305.
Vergne, J., Wittlinger, G., Farra, V., Su, H., 2002. Evidence for upper crustal
anisotropy in the Songpan–Ganze (northeastern Tibet) terrane. Earth Planet.
Sci. 203, 25–33.
Villaseñor, A., Ritzwoller, M.H., Levshin, A.L., Barmin, M.P., Engdahl, E.R.,
Spakman, W., Trampert, J., 2001. Shear velocity structure of central Eurasia from
inversion of surface wave velocities. Phys. Earth Planet. Inter. 123, 169–184.
Wei, W., Unsworth, M., Jones, A., Booker, J., Tan, H., Nelson, D., Chen, L., Li,
S., Solon, K., Bedrosian, P., Jin, S., Deng, M., Ledo, J., Kay, D., Roberts, B.,
2001. Detection of widespread fluids in the Tibetan crust by magnetotelluric
studies. Science 292, 716–718.
Yin, A., Harrison, T.M., 2000. Geologic evolution of the Himalayan–Tibetan
orogen. Annu. Rev. Earth Planet. Sci. 28, 211–280.
Zeyen, H., Fernàndez, M., 1994. Integrated lithospheric modeling combining
thermal, gravity, and local isostasy analysis: application to the NE Spanish
Geotransect. J. Geophys. Res. 99, 18089–18102.
Zeyen, H., Ayarza, P., Fernàndez, M., Rimi, A., 2005. Lithospheric structure
under the western African–European plate boundary: a transect across the
Atlas Mountains and the Gulf of Cadiz. Tectonics 24, TC2001. doi:10.1029/
2004TC001639.
Zhang, Z., Klemperer, S.L., 2005. West-east variation in crustal thickness in
northern Lhasa block, central Tibet, from deep seismic sounding data.
J. Geophys. Res. 110, B09403. doi:10.1029/2004JB003139.
Zhang, P.-Z., Shen, Z., Wang, M., Gan, W., Bürgmann, R., Molnar, P., Wang, Q.,
Niu, Z., Sun, J., Wu, J., Hanrong, S., Xinzhao, Y., 2004. Continuous
deformation of the Tibetan Plateau from global positioning system data.
Geol. Soc. Am. 32, 809–812. doi:10.1130/G20554.1.
Zhao, J., Mooney, W.D., Zhang, X., Li, Z., Jin, Z., Okaya, N., 2006. Crustal
structure across the Altyn Tagh Range at the northern margin of the Tibetan
plateau and tectonic implications. Earth Planet. Sci. Lett. 241, 804–814.
Zhao, W., Mechie, J., Brown, L.D., Guo, J., Haines, S., Hearn, T., Klemperer, S.L.,
Ma, Y.S., Meissner, R., Nelson, K.D., Ni, J.F., Pananont, P., Rapine, R., Ross,
A., Saul, J., 2001. Crustal structure of central Tibet as derived from project
INDEPTH wide-angle seismic data. Geophys. J. Int. 145, 486–498.